void Clock::alarmTime()
{
qDebug()<<"alarmTime";
startNetwork();
readNetwork();
runClock = false;
}
int main( int argc, char ** argv) {
Magick::InitializeMagick(*argv);
readNetwork("base");
int failed = 0, all = 0;
for (int i = 1; i < argc; i++) {
Magick::Image img(argv[i]);
prepareImage(img);
char res = runNetwork(img);
char et = tolower(findChar(argv[i]));
//printf("%c-%c\n", res, et);
if (res != et) failed++;
all++;
}
printf("%d/%d=%.2lf", failed, all, ((double)failed)/all);
}
//gets the file name and reads the *txt file and outputs the readout as a network structure
network read_network_from_file(string file_name){
network net;
//read network from file
ifstream readNetwork(file_name, ios::in);
int v1, v2; //v1 and v2 store the vertices for each edge entry
net.Number_of_Edges = 0; //NumberEdges counts the number of edges as input file streams
vector<vector<int>> preVector;
while (readNetwork >> v1)
{
readNetwork >> v2;
net.Number_of_Edges++;
if (preVector.size() <= max(v1, v2))
{
preVector.resize(max(v1, v2) + 1);
}
preVector[v1].push_back(v2); //add corresponding info for an edge to the to vertices
preVector[v2].push_back(v1);
}
readNetwork.close();
//
net.Number_of_Vertices = preVector.size();
//net.graphVector.resize(net.Number_of_Edges * 2); //graphVector is the finall vector used to represent the network
//net.delimiterInfo.resize(preVector.size() + 1); //the array used to store the information on where in graphVector each array starts
net.dInfo[0] = 0;
int current = 0; //to indicate where in the graphVector we are
for (int i = 0; i < preVector.size(); i++)
{
net.dInfo[i + 1] = net.dInfo[i] + preVector[i].size();
for (int j = 0; j < preVector[i].size(); j++)
{
net.gVector[current] = preVector[i][j];
current++;
}
}
return net;
}
int main()
{
neural::Network net;
char inFile[] = "net1.json";
char outFile[] = "test1.json";
readNetwork(inFile, &net);
writeNetwork(outFile, &net);
/*
std::vector<double> inputValues; //Values to train neural network input with forward-propagation
std::vector<double> targetValues; //Values to train neural network output with back-propagation
std::vector<double> resultValues; //Holds output from neural net
//Train network
myNet.feedForward(inputValues); //Input on input training data
myNet.backPropagation(targetValues); //Train on expected output with backpropagation
myNet.getResults(resultValues);
*/
return 0;
}
